10. Rainfall Runoff Modelling -technic

Question 1

What is a model?

Answer 1

A model is an idealised description of reality or in fact of our perception of reality. It is valid only for certain bounds of application:

• Processes and variables (floods)
• locations
• space scales
• time scales

Question 2

why do we need modelling?

Answer 2

-Limitation of hydrological measurement techniques
-Extrapolation in space (ungauged locations) and time
(future)
-Forecasting, Decision support, planning, impact analysis, research

Question 3

What are the applications areas of RR-Model?

Answer 3

• Design floods: (spliways, flood control reservoirs, input to hydraulic models)
• Planning and impact analysis: land use, climate impact, reservoir management.
• Real time forecast of floods: (reservoir release)

Question 4

Typical RR-model

Answer 4

 HBV-Model (Lindström et al., 1997)
 Model HEC-HMS (Feldmann, 2000)
 Model NASIM (Hydrotec, 2003)
 TOPMODEL (Beven et al., 1995)

Question 5

What are the applications areas of the Water Balance Models?

Answer 5

Assesment of water availability
Agricultural planning (irrigation)
Land use and climate change impacts
Bondary conditions for GW models
Basis for ecohydrological models

Question 6

Typical Water Balance Models

Answer 6

Model ARC-EGMO (Becker et al., 2002)
Model WASIM-ETH (Schulla, 1997)
HBV-Model (Lindström et al., 1997)

Question 7

What are the applications areas of the Ecohydrological Models?

Answer 7

• Assessment of diffuse pollution in ground and surface waters
• Estimation of erosion and sediment transport
• Agricultural planning
• Impact analysis

Question 8

Typical Ecohydrological models:

Answer 8

SWIM (Krysanova et al., 1998)
SWAT (Arnold et al., 1998),
HERMES (Kersebaum, 1995)
CANDY (Franko et al., 1995),

Question 9

Scaling?

Answer 9

spatial or temporal transfer of variables or
parameters between two different scales.
change of temporal or spatial resolution.

Question 10

Regionalisation?

Answer 10

transfer of variables or parameters from known to unknown points in space or time

Question 11

Calibration?

Answer 11

Manual or automatic optimisation of some model parameters.

Maximize simulation performance comparing observed and simulated target variable.

Question 12

Validation?

Answer 12

Test of model performance using a different data set and keeping model parameters constant.

Question 13

Methods to measure the model performance

Answer 13

Mean error or bias -> 0 best
Absolute/Relative standart error -> minimize
Nash-Sutcliff-Criterion-> NSE= 1 best

Question 14

Model uncertainty can result from:

Answer 14

Uncertain knowledge about processes
Uncertain model parameters
Uncertain input variables
Uncertain target/ reference variables

Question 15

How to quantify the uncertainty of a model?

Answer 15

The simplest way is by discussing the performance
measures.
Better is to provide results with confidence bands
(Baye’sche uncertainty estimation, GLUE, different
Monte-Carlo experiments, …)